English translation of the official 2023 Program Implementation Review on the Policy Implications of the Government’s Preference for Interest Subsidies over Direct Loans, issued by the Special Committee on Budget and Accounts. Authored by Junghwan Kim (na32@assembly.go.kr) and combined with complementary sections on guarantee and investment programs prepared by Jisoo Kim (kkjjss@assembly.go.kr). The translation keeps the substance of the Korean report but is edited for clarity and readability.

Original version (Korean) — pp.591–656

1. Introduction

1.A Overview of Public Financing Programs

Public financing programs are policy-based financial tools that channel credit and capital to targeted sectors. They include loans, interest subsidies, credit guarantees, and equity investments.

The Korea Institute of Finance (2013) defines public financing as finance with a public purpose that corrects market failures by improving access and affordability of capital.
The National Assembly Budget Office (2022) describes it as funding and services provided by public financial institutions under the Act on the Management of Public Institutions. Programs are broadly grouped into SME, export, development, housing, and inclusive-finance categories.

Public financing can be classified by funding type:

1. Loan support –
   Direct loans are provided and managed by public institutions that bear the credit risk. Indirect loans are executed through commercial banks, with repayment to the government as agreed.
2. Interest subsidies – The government covers part of the interest-rate gap between policy and market lending.
3. Guarantee support – Institutions such as KODIT and KIBO provide credit guarantees that activate upon borrower default.
4. Equity investment support – The government invests equity through funds or special-purpose vehicles to mobilize private investment.

1.B FY 2023 Budget Execution by Ministry

In 2023, the combined execution of public-financing budgets across 22 ministries was ₩ 43.2 trillion, or 96.3 percent of the ₩ 44.9 trillion final appropriation. Three ministries—Land, Infrastructure and Transport (MOLIT), SMEs and Startups (MSS), and Agriculture, Food and Rural Affairs (MAFRA)—accounted for about 82 percent of the total.

Notes: Includes 22 ministries; totals may not sum due to rounding. Source: Digital Budget and Accounting System (DBAS); ministry submissions (2024).

1.C Expenditure Trends by Support Type, 2019–2023

Between 2019 and 2023, total public-financing outlays averaged around ₩ 45 trillion. Interest-subsidy programs expanded the fastest—growing 20 percent per year and reaching ₩ 1.73 trillion in 2023. Equity-investment programs also increased, while loans stayed broadly stable and guarantees fell as COVID-related support phased out.

Notes: Loan support aggregates items 450-04 and 450-02.
Interest-subsidy data correspond to item 320-05. Guarantee and equity data are compiled from project-level aggregates.
Source: Digital Budget and Accounting System (DBAS).

2. Implementation Review and Findings

This section reviews how loan and interest-subsidy programs were implemented in 2023, focusing on fiscal performance, administrative design, and transparency. It highlights cases of under-execution, over-execution, and emerging risks to fiscal discipline.

2.A Structure of Loan and Interest-Subsidy Programs

The 2023 Budget Compilation and Fund Management Guidelines encouraged ministries to convert direct government loans into interest-subsidy programs. The intent was to lower fiscal outlays while keeping credit accessible.

“Convert direct lending programs into interest-subsidy-based ones, reducing total expenditure while expanding the number of beneficiaries.”

From a budget perspective, direct loans require full disbursement of the principal as expenditure in the year of payment. Interest subsidies, by contrast, fund only the rate difference, easing near-term fiscal pressure. From a policy perspective, direct loans help borrowers with low credit or collateral, while subsidies depend on banks’ willingness to lend and may narrow access for riskier borrowers.

High market interest rates in 2023 also widened the policy-market rate gap, increasing the fiscal cost of subsidies and making some programs less attractive to participants. Several new subsidy-type programs therefore recorded weak performance.

Low Execution after Conversion from Direct Loans

Some programs converted from loans to subsidies in 2022 showed almost no execution in 2023 because of poor coordination and limited demand.
The Ministry of Trade, Industry and Energy (MOTIE) added interest-subsidy lines in its Energy and Resources Special Account and Electricity Industry Fund, yet none were implemented.

Notes: Programs created through the 2022 supplementary budget to replace loan funding.
Source: Ministry of Trade, Industry and Energy (MOTIE).

MOTIE attributed the zero-execution outcome to high interest rates and to banks’ preference for risk-free direct loans. The case illustrates the need for feasibility checks and coordination with financial institutions before structural changes.

Over-Execution and Fiscal Risks

Other ministries exceeded approved ceilings, creating unpaid obligations and potential breaches of parliamentary budget authority. Two examples are the Agricultural Loan Interest Subsidy Program under MAFRA and the Forest Management Composite Fund under the Korea Forest Service (KFS).

Notes: Adjustment reflects a ₩ 481 million transfer for disaster-recovery activities. Unpaid = carry-over obligations to FY 2024 under the Special Account for Agricultural and Rural Structure Improvement.
Source: Ministry of Agriculture, Food and Rural Affairs (MAFRA).

The unpaid balance of ₩ 162 billion (about 32 percent of the 2024 budget) shows the quasi-mandatory nature of such subsidies and the fiscal strain that arises when appropriations fall short.

Notes: KFS reported total subsidy needs exceeding the approved budget by ₩ 1.53 billion.
Source: Korea Forest Service (2024).

Repeated over-execution of quasi-obligatory interest subsidies weakens fiscal discipline and can generate arrears. Stronger monitoring and reconciliation procedures are required.

Oversight by Type of Interest-Subsidy Support

Government interest-subsidy programs differ by (i) how support is paid to financial institutions and (ii) how the borrower’s rate is set.

• Payment to financial institutions 1) Actual-cost reimbursement – government reimburses the measured interest gap;
  2) Estimated lump-sum (advance) – government pays a pre-agreed rate × average loan balance.

• Setting the borrower rate 1) Floating-subsidy model – borrower rate is fixed (target), subsidy varies by case;
  2) Fixed-subsidy model – subsidy rate is fixed, borrower rate varies.

Notes: FSC = Financial Services Commission; MAFRA = Ministry of Agriculture, Food and Rural Affairs; MOLIT = Ministry of Land, Infrastructure and Transport. Program names as used in ministry submissions (2024).
Source: Authors’ classification based on program documents.

In reimbursement schemes with a fixed borrower target rate (e.g., 2.5 percent), the government pays the difference between each bank’s actual lending rate and the target. This structure weakens incentives for both parties to lower pre-subsidy rates because borrowers face the same final rate regardless of shopping around, and banks may benefit from providing higher pre-subsidy quotes.

For FSC Interest-Subsidy Support in 2023, ₩56.82 billion out of ₩63.02 billion was executed, corresponding to an execution rate of 90.2 percent. The program consisted of two subcomponents: the Low-Income Small Business Interest Subsidy (budget ₩27.03 billion; execution ₩21.88 billion) and the Hope Plus Interest Subsidy (budget ₩35.99 billion; execution ₩35.01 billion). The Low-Income Small Business Interest Subsidy operated as a reimbursement scheme tied to fixed borrower target rates, whereas the Hope Plus Interest Subsidy adopted a lump-sum structure.

Across participating banks in the Low-Income Small Business Interest Subsidy program, the applied subsidy rates differed at the time of execution.

Note: Borrower target rate = 2.5%.
Source: Financial Services Commission (FSC).

Recalculating each bank’s pre-subsidy lending rate, given that the subsidy rate corresponds to 70 percent of the rate gap relative to the 2.5 percent target, allows for a simple cross-bank comparison. This reveals a notable dispersion in the bank-calculated pre-subsidy rates, suggesting heterogeneity in pricing practices across institutions.

Notes: The illustrative government subsidy rate assumes a 70% subsidy on the interest rate differential. Source: FSC; author’s presentation.

Accordingly, when interest-subsidy programs are operated by linking the subsidy rate to the target lending rate on a loan-by-loan basis, it is essential to enforce strict oversight to prevent moral hazard by individual financial institutions. In addition, close scrutiny is required to assess the appropriateness of subsidy claims from institutions whose average calculated lending rates are higher than those of their peers.

In lump-sum interest-subsidy programs, the subsidy is calculated as a pre-agreed rate (e.g., 2.1 percent) applied to the average outstanding loan balance. This structure requires ex-post settlement to reconcile differences between the assumed and actual lending rates and balances.

For the FSC Hope Plus Interest Subsidy (2023), the budget was based on an assumed average lending rate of 6.3 percent, a target rate of 3.3 percent, and a 70 percent coverage ratio, yielding a subsidy rate of 2.1 percent. The assumed average lending rate used for budgeting increased from 4.5 percent in 2022 to 6.3 percent in 2023.

Following this scheme, KODIT disbursed lump-sum subsidies to 14 banks:

Note: Lump-sum payments based on pre-agreed rate applied to reported balances and days outstanding.
Source: FSC/KODIT.

As of July 2024, there had been no ex-post reconciliation against realized 2023 lending rates. Because the weighted-average rate in the low-income program was 5.61 percent, it is likely that the Hope Plus program, which serves higher-credit borrowers, recorded an average rate lower than the assumed 6.3 percent. In the absence of proper settlement, fiscal costs may therefore be overstated.

Accordingly, annual reconciliation should be required for lump-sum programs. This includes updating realized average lending rates and balances, clawing back any overpayments, and publicly disclosing settlement results.

2.B. Strengthening Management and Operational Frameworks

Program management is split between (i) indirect cost inclusion (administrative costs absorbed in the operating budgets of affiliated agencies) and (ii) direct budget allocation (explicit management lines). Audits identified two weaknesses: some agencies retain a share of loan repayments as fees (off-budget), and some deduct administrative costs from interest-subsidy envelopes (non-transparent).

Source: Ministry submissions (2024).

2.B.1 Off-Budget Retention of Loan Repayments

Several programs allow implementing agencies to directly retain portions of principal/interest as handling fees—an exception to gross-budgeting.

Source: Ministry submissions (2024).

Policy implication. Replace off-budget retention with appropriated management lines; publish fee formulae and rates; and reconcile annually.

2.B.2 Deducting Administration from Interest-Subsidy Envelopes

Some programs draw administrative costs directly from subsidy funds, which are not subject to detailed ex-post reconciliation.

Source: Ministry data (2024).

Policy implication. Prohibit charging administration costs to item 320-05 (interest-subsidy) and fund these costs through explicit budget lines with settlement requirements.

2.B.3 Delegation and Re-outsourcing without Clear Legal Basis (MSIT/IITP)

Under the ICT Convergence Technology Development Loan Support, IITP subcontracted fund management to IBK and allowed IBK to deduct a 0.10% fee directly from repayments, while commercial banks deducted additional handling (0.6%) and guarantee (0.9%) fees before remitting.

Program statutes (Framework Act and Enforcement Decree) permit delegation to sector agencies (e.g., Korea Communications Agency) and operation “through financial institutions,” but do not explicitly authorize re-outsourcing of fund management to commercial banks or direct deduction of fees from loan proceeds.

Policy implication. Avoid allowing commercial banks to use loan principal and interest payments to cover management or administrative fees that effectively function as private-delegation service charges, especially in the absence of a legal basis for re-delegation.

2.B.4 Fee-Rate Governance and Cross-Agency Comparison

Loan handling fees vary across ministries and programs; many anchor at 1.0%, but several use higher rates (e.g., 1.5%). A central benchmark and periodic review are warranted.

Source: Ministry submissions (2024).

Case examples.

• MOEF Self-Reliance Loans for Persons with Disabilities (2023): fee 1.5%
• MOTIE loan programs: several operate near 1.5% handling fees (Urban Gas Facility, Renewable Energy, Green Innovation).

Policy implication. Establish unified, cross-ministerial standards for loan-handling fee rates, and align ministry-specific regulations with a common framework.

2.B.5 Settlement of Unused Subsidy Balances and Interest Income (MOTIE/KIAT)

In 2023, MOTIE disbursed ₩2.49 bn in interest-subsidy funds to KIAT for EV supply-chain support. Only ₩853 mn was executed for new cases; ₩1,549 mn remained with KIAT without formal carryover approval. Interest income accrued on idle balances (₩15.61 mn in 2022; ₩68.25 mn in 2023) had not been remitted to the treasury by end-June 2024.

Source: MOTIE; KIAT (2024).

Source: MOTIE; KIAT (2022–2023).

Policy implication. Ensure timely restitution of any interest income generated from delayed execution of interest-subsidy funds by KETEP, and return such proceeds to the national treasury without delay.